Stem cell heterogeneity drives the parasitic life cycle of Schistosoma mansoni

15311327048536 1..23Attribution License, which
credited.
Stem cell heterogeneity drives the parasitic life cycle of Schistosoma mansoni Bo Wang1,2,3*, Jayhun Lee3†‡, Pengyang Li1†, Amir Saberi3†§, Huiying Yang1, Chang Liu4, Minglei Zhao4, Phillip A Newmark3‡*
1Department of Bioengineering, Stanford University, Stanford, United States; 2Department of Developmental Biology, Stanford University School of Medicine, Stanford, United States; 3Department of Cell and Developmental Biology, Howard Hughes Medical Institute, University of Illinois at Urbana-Champaign, Urbana, United States; 4Department of Biochemistry and Molecular Biology, University of Chicago, Chicago, United States
Abstract Schistosomes are parasitic flatworms infecting hundreds of millions of people. These
parasites alternate between asexual reproduction in molluscan hosts and sexual reproduction in
mammalian hosts; short-lived, water-borne stages infect each host. Thriving in such disparate
environments requires remarkable developmental plasticity, manifested by five body plans
deployed throughout the parasite’s life cycle. Stem cells in Schistosoma mansoni provide a
potential source for such plasticity; however, the relationship between stem cells from different life-
cycle stages remains unclear, as does the origin of the germline, required for sexual reproduction.
Here, we show that subsets of larvally derived stem cells are likely sources of adult stem cells and
the germline. We also identify a novel gene that serves as the earliest marker for the schistosome
germline, which emerges inside the mammalian host and is ultimately responsible for disease
pathology. This work reveals the stem cell heterogeneity driving the propagation of the
schistosome life cycle.
DOI: https://doi.org/10.7554/eLife.35449.001
Introduction Flatworms include more than 44,000 parasitic species that form one of the largest groups of meta-
zoan endoparasites (Loker and Hofkin, 2015). Their life cycles typically involve asexually and sexu-
ally reproducing stages, each with its own distinct body plan and strategy to enhance transmission
between multiple hosts (Clark, 1974; Pearce and MacDonald, 2002; Viney and Cable, 2011).
Although the life cycles of these parasites were established more than a century ago, they have only
recently been studied in cellular and molecular terms (Matthews, 2011). Since many parasitic flat-
worms are pathogenic, their life cycles are also the routes for disease transmission (Hoffmann et al.,
2014). Therefore, a deeper understanding of these life cycles is significant from both basic science
and medical perspectives, as blocking transmission is an effective approach to fighting parasitic
diseases.
Focusing on the cells that may drive such parasitic life cycles, we study Schistosoma, a parasitic
flatworm infecting over 250 million people, which causes the major neglected tropical disease, schis-
tosomiasis (Hoffmann et al., 2014). Schistosomes are transmitted through snail intermediate and
human definitive hosts. Their life cycle begins with the parasite egg excreted from the mammalian
host into water, releasing a free-swimming miracidium larva. The miracidium penetrates a snail host
and transforms into a mother sporocyst that undergoes asexual clonal expansion to produce many
daughter sporocysts that leave the mother and colonize other snail tissues. These daughters either
self-renew to produce more daughters or enter embryogenesis to produce infective cercariae
Wang et al. eLife 2018;7:e35449. DOI: https://doi.org/10.7554/eLife.35449 1 of 23
RESEARCH ARTICLE
2009). This cloning process is repeated, allowing massive numbers of cercariae to be produced from
a single miracidium. Mature cercariae emerge from the snail into water, then burrow through the
skin of a mammalian host to become schistosomula. This transition initiates the sexual portion of the
life cycle. Schistosomula then migrate to species-specific niches in the host vasculature and develop
into juvenile worms (Basch, 1991; Wilson, 2009). Juveniles remodel their tissues extensively to build
a functional digestive system, and after they begin feeding on host blood, undergo massive growth
and develop sexual reproductive organs de novo (Clegg, 1965). Male and female worms pair to pro-
duce fertilized eggs, which are excreted to continue the life cycle.
A long-standing hypothesis proposes that a lineage of totipotent stem cells, called ‘germinal
cells’, persists throughout the schistosome life cycle and drives reproduction (Cort et al., 1954;
Clark, 1974; Whitfield and Evans, 1983). Histological and ultrastructural studies defined these cells
in miracidia and sporocysts by their stem cell-like morphology and rapid proliferation (Schutte, 1974;
Pan, 1980). Recently, we showed that these germinal cells indeed drive proliferation within develop-
ing sporocysts and share some molecular signatures with stem cells from diverse organisms
(Wang et al., 2013). In contrast, the cellular source of the schistosome germline, which underlies
sexual reproduction in the mammalian host, remains an open question. Furthermore, because
somatic stem cells were only recently identified in adult schistosomes (Collins et al., 2013), the rela-
tionships between germinal cells, germ cells, and somatic stem cells are unclear.
To clarify these relationships, we transcriptionally profiled stem cells from Schistosoma mansoni
asexual (sporocyst) and sexual (juvenile) stages at both population and single-cell levels. We identi-
fied four transcriptionally distinct populations and validated this heterogeneity by in situ hybridiza-
tion. By characterizing the behavior of these stem cells at major developmental transitions, we found
that larvally derived stem cells serve as the source for the parasite’s adult stem cells. We also identi-
fied a novel gene that is activated during development inside the mammalian host and serves as the
eLife digest Parasitic flatworms called schistosomes infect around 250 million people, causing
the disease schistosomiasis. Schistosomes live complex lives, spending part of their life cycle inside
snails and part of it inside mammals; short-lived, water-borne stages infect each of these hosts. To
thrive in such different environments, schistosomes go through several life-cycle stages. At each
stage the flatworms transition to a new body plan adapted to its new environment. Understanding
how these transitions occur could help researchers devise new strategies for eliminating these
parasites.
Previous research suggested that stem cells help schistosomes transition to new body plans.
Stem cells have the ability to transform into many different cell types, and have been found in
schistosome larvae and adults. However, the relationship between the larval and adult stem cells
was not clear.
Wang et al. used transcriptional profiling, a technique that measures the genes currently in use in
different cells, to study the stem cells in the schistosome species Schistosoma mansoni. This
uncovered four types of stem cell, each of which uses a slightly different combination of genes.
Examining the behaviour of these cells at different schistosome life-cycle stages revealed that
certain larval stem cells produce adult stem cells. Other larval stem cells seem to be the source of
the ‘germline’ cells that make gametes (egg and sperm) and allow the parasites to reproduce
sexually.
Schistosomes only produce germline cells when they are inside mammals. Wang et al. found that
as juvenile flatworms develop inside mouse blood vessels, a gene called eledh becomes active in
some of their stem cells. Further investigation showed that this activity is the earliest indicator that
germline cells are developing and is also required for proper development of the germline. This
knowledge, along with future work to characterize the roles of the stem cell populations identified
by Wang et al., could ultimately help researchers develop new ways to stop the spread of
schistosomiasis.
Research article Microbiology and Infectious Disease Stem Cells and Regenerative Medicine
ing the development and propagation of these important parasites.
Results
Single-cell RNAseq defines three major sporocyst stem cell classes Each miracidium carries 10–20 germinal cells (Pan, 1980; Cort et al., 1954; Wang et al., 2013),
which expand massively and differentiate to produce many daughter sporocysts (Figure 1A, and
Figure 1—figure supplement 1). Our recent work has shown that germinal cells exhibit heterogene-
ity within this population (Wang et al., 2013), revealed by the distinct proliferation kinetics and
expression of a schistosome homolog of nanos (Wang and Lehmann, 1991), a conserved regulator
of germ cell development (Juliano et al., 2010; Wang et al., 2007) also expressed in the schisto-
some adult stem cells (Collins et al., 2013). To characterize this heterogeneity further, we isolated
and transcriptionally profiled these stem cells from in vitro-transformed mother sporocysts
(Figure 1B).
Principal component analysis (PCA) of single-cell transcriptomes revealed three major cell classes
(Figure 1C). We designated these classes based upon their respective markers: k-cells (kappa indi-
cates klf+nanos-2+); j-cells (phi indicates fgfrA,B+); and d-cells (delta indicates double-positive for
nanos-2 and fgfrA,B). The difference between k and d/j-cells extends along PC1, and contributes
to ~30% of the total variance among cells, whereas the difference between d and j-cells is second-
ary, delineated by PC2 and contributing ~10% of the total variance. For example, nanos-2 exhibits
almost equal loadings on both PCs, negative on PC1, positive on PC2, consistent with its expression
in both k and d-cells. Based on projections along the first two PCs (Treutlein et al., 2014), we identi-
fied additional genes that contribute to the distinctions between classes: a schistosome p53 homo-
log and a zinc finger protein (zfp-1) expressed abundantly in d-cells and at lower levels in j-cells;
and a hes family transcription factor (hesl) expressed specifically in j-cells (Figure 1D and E,
Supplementary file 1). We validated these transcriptomic findings by fluorescent in situ hybridiza-
tion (FISH) on in vitro-cultured mother sporocysts (Figure 1—figure supplement 2). Unfortunately,
the k class-specific marker klf was expressed at very low levels (Figure 1E), beneath the detection
limits of our current FISH protocol.
In addition to these class-defining genes, the divergence of the three cell classes is manifested by
hundreds of other genes that exhibit various levels of statistically significant differences between
classes (Figure 1—figure supplement 3). However, these genes comprise only a small fraction of
transcripts detected in these cells (N = 6,661), and most of them are not enriched in stem cells com-
pared to differentiated cells. Notably, very few transcripts are specific to individual cell classes, with
j-cells showing the fewest specific markers. These observations confirm that sporocyst stem cells,
regardless of the subpopulation to which they belong, share a common transcriptomic profile.
Stem cell classes display distinct spatiotemporal patterns throughout asexual development Examining fgfrA and nanos-2 enabled us to distinguish all three cell classes in situ: j-cells express
fgfrA, k-cells express nanos-2, and d-cells express both. Thus, we followed these cells throughout
intramolluscan development by monitoring fgfrA and nanos-2 expression. After the first week of
infection, asexually produced embryos– identified as compact, spherical cell clusters (Schutte, 1974)
and from which daughter sporocysts will arise– begin to develop (Figure 2A). j-cells were distrib-
uted beneath the parasite’s outer layer and excluded from daughter embryos. d-cells were found in
large clusters within embryos. k-cells clustered with d-cells in embryos and were found in extraem-
bryonic tissues as singlets or doublets, suggested to be the source of developing embryos in previ-
ous histological studies (Schutte, 1974).
Two weeks post-infection, mother sporocysts contain many mature daughter sporocysts that are
ready to leave the mother and migrate elsewhere in the snail. At this stage k-cells comprised the
vast majority (>85%) of stem cells in these daughters (Figure 2B); fewer d-cells were observed and
j-cells were mostly excluded. However, one week later, when post-migratory daughters colonized
new regions of host tissue (Figure 2C), all three classes reappeared as intermingled populations,
consistent with k-cells generating the other stem cell types.
-8
0
8
16
-0.4
-0.2
0.0
0.2
0.4
zfp-1 hesl klf
φ δ κ φ δ κ φ δ κ φ δ κ
lo g
extraembryonic space
tail bud
penetration glands
mature cercaria
A migration
mature cercaria
eggs juvenilesadults
blood vessel
Figure 1. Single-cell RNAseq reveals stem cell classes in sporocysts. (A) Schematic of the schistosome life cycle.
Images depicting developmental stages shown in Figure 2 are labeled accordingly. (B) Dissociated cells were
gated using forward scattering (FSC), side scattering (SSC), and DyeCycle Violet (DCV) fluorescence to isolate S or
G2/M phase cells from mother sporocysts. Dead cells and debris (<30% of total events) were pre-excluded based
on high TOTO-3 fluorescence. Right: Sorted G2/M phase cells from mother sporocysts visualized by DIC and
fluorescence microscopy. (C) PCA of 35 single-cell transcriptomes of sporocyst stem cells. Summative variances are
reported in percentages. Assignment of cell classes is based on hierarchical clustering. (D) Selected genes with
heavy loadings are plotted in projection on the first 2 PCs. The projection on each axis represents the correlation
coefficient of the respective gene with each principal component. (E) Box plots of expression levels of selected
class-dependent genes. ago2-1 expression is also shown as a ubiquitous stem cell marker. Boxes indicate quartiles
Figure 1 continued on next page
Intramolluscan development culminates with the production of infectious cercariae. In early cer-
carial embryos (dashed circle in Figure 2D–2E), j-cells were found concentrated both anteriorly and
posteriorly, where the mouth and tail bud form, respectively. Additionally, two clusters of k-cells
were observed posterior to the penetration glands (Figure 2D–2E), at the site of the germinal cell
cluster, considered gonadal primordia based on histological and ultrastructural studies (Cheng and
Bier, 1972; Dorsey et al., 2002). In the mature cercarial body (Figure 2F–2G), the k-cell pair poste-
rior to the glands expands into two clusters that contain multiple cells each. In parallel, five d-cells
were detected in a regular pattern around the penetration glands, with one at the midline and two
pairs laterally (Figure 2E), whereas j-cells are absent at this stage. Since the cercarial body (but not
the tail) penetrates the mammalian host, only d and k-cells, but not j-cells, may be passed to the
intramammalian (sexual) stage.
Larvally derived stem cells drive initial proliferation in schistosomula After emerging from the snail into water, cercariae burrow through mammalian host skin and their
bodies transform into the next life-cycle stage, the schistosomula. At this stage, the parasites do not
grow for several weeks, until they reach the hepatic portal vein. Thus, the extent of proliferation in
the initial days after infection has been unclear (Clegg, 1965), with mitotic cells only detected 4 days
post-infection (Clegg and Smithers, 1972). Furthermore, because the adult stem cells have only
been identified recently (Collins et al., 2013), their developmental origin has yet to be investigated.
The identification of d and k-cells in cercariae provides a potential source of new multipotent cells
upon entry into the mammalian host.
We mimicked this transition by exposing cercariae to ex vivo mouse tail-skin biopsies and collect-
ing transformed schistosomula on the other side of the skin (Clegg and Smithers, 1972;
Protasio et al., 2013). Following skin penetration, we assayed proliferation in schistosomula via EdU
labeling (Figure 3A). Between 22 and 36 hr post-transformation, we observed five EdU+ cells around
the penetration glands, anterior to the ventral sucker (Figure 3B), and confirmed that they were d-
cells (fgfrA+nanos-2+) (Figure 3C). During the next 12 hr, these cells completed mitosis, indicated by
the appearance of five EdU+ doublets (Figure 3B). Thereafter, the number of EdU+ nuclei steadily
increased (Figure 3D), but proliferation was restricted anteriorly to the ventral sucker, until one
week later, when two clusters of ~2–3 EdU+ cells appeared in the ‘germinal cell cluster’ region pos-
terior to the ventral sucker, where k-cells are found (Figure 3B, Days 9–10). EdU+ cells were not
detected in irradiated worms (Figure 3E), consistent with previous reports that irradiation leads to
developmental defects and reduced pathogenicity (Wilson, 2009). These results suggest that a
small, fixed number of k and d-cells are transmitted to the mammalian host. Because these cells
appear to be the only dividing cells in schistosomula, they are likely the source of the recently identi-
fied stem cells in adult schistosomes (Collins et al., 2013; 2016).
Stem cells in juveniles reveal germline and somatic populations Intramammalian growth initiates after schistosomula migrate into the portal vein, around 2 weeks
post-infection (Clegg, 1965; Basch, 1981). To characterize proliferation driving juvenile growth, we
harvested EdU-labeled parasites 3 weeks post-infection. Worms displayed a range of sizes based
upon differences in arrival time, enabling a developmental time course to be reconstructed from a
Figure 1 continued
and medians, whiskers show maxima and minima, and dots represent outliers (above and below 1.5X interquartile
range). *p<0.01 (t-test). p-values were estimated based on multiple models using either TPM or log2(TPM +1) as
expression values.
The following figure supplements are available for figure 1:
Figure supplement 1. S. mansoni sporocysts develop in B. glabrata snails.
Figure supplement 2. Distribution of stem cell classes in in vitro-transformed mother sporocysts.
Figure supplement 3. Number of genes differentially expressed between stem cell classes.
Figure 2. Stem cells exhibit class-specific spatial and temporal patterns during intramolluscan development. (A)
FISH of nanos-2 and fgfrA reveals spatial distributions of cell classes in a mother sporocyst containing daughter
embryos (dashed circles), 10 days post-infection (dpi). sWGA (grey) labels the parasite surface (tegument). Arrows:
extraembryonic k-cells. Right: rendered image of that shown to the left. Spheres: cell centers; yellow arrowheads:
f-cells beneath the surface. (B) Mature daughter sporocysts contained in a mother sporocyst 15 dpi. Dashes
outline the body surfaces of daughters. In daughters, k-cells comprised 522 out of 592 counted stem cells in these
daughters, while fewer d-cells (61/592) and j-cells (9/592) were observed. (C) Daughter sporocysts in snail
ovotestis, 25 dpi. (D) Early cercarial embryo before the tail bud forms. j-cells were found concentrated both
anteriorly and posteriorly, where the mouth and tail bud form; only two k-cells were detected posterior to the two
penetration glands. (E) Later development during cercarial embryogenesis. j-cells were mostly located in the tails;
the k-cell pair were posterior to the glands. (F) Mature cercarial body in daughter sporocysts 30 dpi (dorsal view).
1: anterior midline cell; 2L, 2R: anterior lateral cells; 3L, 3R: posterior lateral cells. Arrowheads: k-cells, asterisks: f-
cells, arrows: d-cells. sWGA: penetration glands. Note that only the cercarial body is transmitted to the mammalian
host, whereas tails are discarded outside of the host epidermis during penetration. (G) k-cell clusters, magnified
from (F), contain multiple cells each (single confocal section). Arrows: individual nuclei in k-cell clusters. Scale bars:
20 mm. All images are maximum intensity projections from 30 mm tissue cryosections. Since animals are thicker
than the sections, parasite surface and penetration glands were used to determine the orientation and position of
the sections.
https://doi.org/10.7554/eLife.35449.007
https://doi.org/10.7554/eLife.35449
1
1
2
0-36hrs
1
2R
3R
3L
2L
E
0
10
20
30
# E
i
EdU (h) =3 8 12 16 22 36 48 64 24 (D9–D10)
20 20 20 20 20 20 2613 18
D
N =
2R
3R
1
3L
2R
3R
1
3L
nanos-2
fgfrA
EdU
DAPI
Figure 3. Larvally derived stem cells proliferate in schistosomula. (A) Schematic of in vitro transformation from
cercariae to schistosomula and EdU labelling. (B) EdU+ cells are detected medially (1) and laterally (2L, 2R, 3L,3R)
at the locations of d-cells in cercarial bodies (ventral view). These cells divide to generate doublets, indicated by
arrows. The time of EdU pulse post-transformation is indicated. Images are maximum intensity projections of
confocal stacks. (C) Confocal maximum intensity projection of FISH of nanos-2 and fgfrA on schistosomula at 2
days post-transformation confirms that only d-cells incorporated EdU. Right: magnified images of boxed cells. (D)
Quantification of EdU incorporation after transformation; x axis: length of EdU treatment post-transformation.
Means and standard deviations are specified. N: number of worms analyzed. (E) Irradiated cercariae exhibit no
EdU+ cells after transformation, which confirms that EdU specifically labels proliferating cells.
static time point. EdU labeling revealed a posterior growth zone (PGZ) that extended as the para-
sites grew (Figure 4A). In…

Stem cell heterogeneity drives the parasitic life cycle of Schistosoma mansoni

Documents

parasitic life cycle